CN114245281A - Voice performance test method and system - Google Patents

Abstract

The invention discloses a voice performance testing method and a system, comprising the following steps: acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested; acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal; acquiring the microphone acquisition signal and the algorithm processing signal, and inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index; comparing the voice performance index with a voice performance threshold value to generate a voice performance test result; the method and the device have the effect of improving the efficiency of the voice equipment during batch testing.

Description

Voice performance test method and system

Technical Field

The invention relates to the technical field of voice performance testing, in particular to a voice performance testing method and system.

Background

At present, when a voice performance test is performed on voice equipment, noise audio or test audio needs to be played by means of external equipment, and the audio recorded by the equipment to be tested needs to be transmitted to an electronic computer and compared and analyzed by professional voice performance test software on the electronic computer.

The existing voice equipment may carry various different operating systems, the playing/recording formats of the existing voice equipment are various and poor in compatibility, format conversion is needed when professional voice performance testing software is introduced for voice performance testing, and the existing voice equipment is complex in operation and low in efficiency.

For the above related art, the inventor thinks that there is a problem that the efficiency is low due to the complicated voice device performance test procedure.

Disclosure of Invention

In order to improve the efficiency of voice equipment in batch test, the application provides a voice performance test method.

The first objective of the present application is achieved by adopting the following technical scheme:

a voice performance testing method comprises the following steps:

acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested; acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal;

acquiring the microphone acquisition signal and the algorithm processing signal, and inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

and comparing the voice performance index with a voice performance threshold value to generate a voice performance test result.

By adopting the technical scheme, the test item information is obtained so as to obtain the loudspeaker and the audio frequency which are required to be used by the next test item from the test item information, when the test is started, an audio playing instruction is generated based on the loudspeaker and the audio frequency which are required to be used by the test item and is sent to the external loudspeaker and the built-in loudspeaker of the equipment to be tested, an audio recording instruction is generated and is sent to the built-in microphone of the equipment to be tested, so that the equipment to be tested plays the test audio frequency, and a microphone collecting signal is recorded; inputting signals collected by a microphone into an audio processing model so as to obtain algorithm processing signals processed by various algorithms in the audio processing model; the microphone acquisition signal and the algorithm processing signal are input into a built-in voice performance evaluation model of the equipment to be tested for comparison and analysis so as to obtain a voice performance index reflecting the voice processing performance of the audio processing model, the microphone acquisition signal and the algorithm processing signal acquired by the equipment to be tested are not required to be transmitted to an electronic computer for testing through professional voice performance testing software, the purchase cost of the professional voice performance testing equipment and the professional voice performance testing software is saved, and the problem of low testing efficiency caused by compatibility is solved; and comparing the voice performance index with a preset voice performance threshold value to generate a voice performance test result for evaluating the voice processing performance of the audio processing model.

The present application may be further configured in a preferred example to: before the step of obtaining the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation, and obtaining a voice performance index, the method further comprises the following steps:

acquiring a noise reduction evaluation algorithm, an echo cancellation evaluation algorithm and a duplex echo cancellation evaluation algorithm, and creating a voice performance evaluation model based on the noise reduction evaluation algorithm, the echo cancellation evaluation algorithm and the duplex echo cancellation evaluation algorithm;

and sending the voice performance evaluation model to the equipment to be tested.

By adopting the technical scheme, a voice performance evaluation model is created, wherein the voice performance evaluation model comprises a noise reduction evaluation algorithm, an echo cancellation evaluation algorithm and a duplex echo cancellation evaluation algorithm which are respectively used for evaluating the noise reduction performance, the echo cancellation performance and the duplex echo cancellation performance of the voice processing model on voice processing; the voice performance evaluation model is sent to the equipment to be tested, so that the equipment to be tested can directly compare a microphone collecting signal recorded by the equipment to be tested with an algorithm processing signal generated by processing through the audio processing model, the problem of compatibility possibly caused by the fact that audio needs to be transmitted to other equipment for processing in the traditional voice performance test is solved, meanwhile, the processor performance of the equipment to be tested is effectively utilized, and the cost for purchasing the voice performance test equipment is reduced.

The present application may be further configured in a preferred example to: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and creating an energy value waveform display window, displaying energy value waveforms of the microphone acquisition signal and the algorithm processing signal in the energy value waveform display window, and comparing the waveforms of the microphone acquisition signal and the algorithm processing signal in real time.

By adopting the technical scheme, the energy value waveform display window is created, and the energy value waveforms of the microphone acquisition signals and the algorithm processing signals are displayed in the energy value waveform display window for comparing the difference between the microphone acquisition signals recorded by the equipment to be tested and the algorithm processing signals processed by the audio processing model in real time, so that testers can more visually and conveniently see the test effect, the problems occurring in the voice performance test process can be conveniently eliminated in time, and the detection accuracy is improved.

The present application may be further configured in a preferred example to: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and acquiring the microphone acquisition signal and the noise reduction algorithm processing signal, aligning the microphone acquisition signal and the noise reduction algorithm processing signal, and calculating a voice suppression ratio and a noise suppression ratio as voice performance indexes.

By adopting the technical scheme, the phase alignment is carried out on the microphone acquisition signal and the noise reduction algorithm processing signal so as to prevent the problem that the phase difference between the microphone acquisition signal and the noise reduction algorithm processing signal causes mutual cancellation of sound signals, the voice suppression ratio and the noise suppression ratio are calculated to be used as voice performance indexes so as to evaluate the voice suppression effect and the noise suppression effect of the noise reduction algorithm, and the lower the voice suppression ratio and the higher the noise suppression ratio, the better the noise reduction performance of the noise reduction algorithm is.

The present application may be further configured in a preferred example to: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and acquiring the microphone acquisition signal and the echo cancellation algorithm processing signal, aligning the microphone acquisition signal and the echo cancellation algorithm processing signal, and calculating an echo return loss gain value as a voice performance index.

By adopting the technical scheme, the phase alignment is carried out on the microphone acquisition signal and the echo cancellation algorithm processing signal so as to prevent the problem that the acoustic signals are mutually offset due to the phase difference between the microphone acquisition signal and the echo cancellation algorithm processing signal, the echo return loss gain value is calculated to be used as a voice performance index so as to evaluate the echo cancellation effect of the echo cancellation algorithm, and the echo cancellation performance of the echo cancellation algorithm is better if the echo return loss gain value is larger.

The present application may be further configured in a preferred example to: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

respectively acquiring the microphone acquisition signal and the duplex algorithm processing signal, aligning the microphone acquisition signal and the duplex algorithm processing signal, subtracting the microphone acquisition signal from the duplex algorithm processing signal to obtain an algorithm evaluation signal, aligning the microphone acquisition signal and the algorithm evaluation signal, and calculating a duplex voice suppression ratio and a duplex echo suppression ratio as voice performance indexes.

By adopting the technical scheme, the phase alignment is carried out on the microphone acquisition signal and the duplex algorithm processing signal so as to prevent the problem that the sound signals are mutually offset due to the phase difference between the microphone acquisition signal and the duplex algorithm processing signal, the microphone acquisition signal is subtracted from the duplex algorithm processing signal to obtain an algorithm evaluation signal, then the phase alignment is carried out on the microphone acquisition signal and the algorithm evaluation signal, a duplex voice suppression ratio and a duplex echo suppression ratio are calculated as voice performance indexes so as to evaluate the voice suppression effect and the echo suppression effect of the duplex echo cancellation algorithm, and the duplex echo cancellation performance of the duplex echo cancellation algorithm is better when the duplex voice suppression ratio is lower and the duplex echo suppression ratio is higher.

The present application may be further configured in a preferred example to: the method comprises the following steps of obtaining test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the device to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the device to be tested, and the method also comprises the following steps: acquiring a test serial number and position information of equipment to be tested, and sending a test preparation instruction to the motor based on the test serial number and the position information of target equipment to be tested;

the method comprises the steps of obtaining position information of target equipment to be tested in real time, and sending a test starting instruction to an equipment bearing wheel and an external loudspeaker when the position information of the target equipment to be tested is successfully matched with a preset target position.

By adopting the technical scheme, the test serial numbers are set for all the devices to be tested, the position information of the devices to be tested is obtained, a test preparation instruction is sent to the motor based on the test serial numbers and the position information of the target devices to be tested, and motor action parameters for moving the target devices to be tested to the preset target position are provided for the motor; the method comprises the steps of acquiring position information of target equipment to be tested in real time, and sending a test starting instruction to the target equipment to be tested and an external loudspeaker to test the voice performance of the target equipment to be tested when the target equipment to be tested is located at a preset target position.

The present application may be further configured in a preferred example to: after the step of comparing the voice performance index with the voice performance threshold value and generating a voice performance test result, the method also comprises the following steps:

and after the target equipment to be tested is tested, marking the next target equipment to be tested from the equipment to be tested based on the test serial number of the tested target equipment to be tested.

By adopting the technical scheme, after the test of one target device to be tested is finished, the next target device to be tested is selected from the devices to be tested and marked based on the test serial numbers of the target device to be tested and the rest devices to be tested, so that the voice performance testing device can test the next target device to be tested conveniently, and the effect of continuously and automatically testing a plurality of devices to be tested is achieved.

The second objective of the present application is achieved by adopting the following technical scheme:

a voice performance testing device comprises a base, an equipment bearing wheel and an external loudspeaker, wherein a motor is arranged on the base, an output shaft of the motor is coaxially and fixedly connected to the equipment bearing wheel, a plurality of equipment fixing components used for fixing equipment to be tested to the equipment bearing wheel are arranged on the equipment bearing wheel, a signal transmitter used for mutually transmitting signals with the equipment to be tested is also arranged on the equipment bearing wheel, and the external loudspeaker is arranged above the equipment bearing wheel; the motor is provided with a first controller used for receiving a test preparation instruction in the voice performance test method to control an output shaft of the motor to rotate, the equipment bearing wheel is provided with a second controller used for receiving a test starting instruction in the voice performance test method to control the signal transmitter to send the test starting instruction to the target equipment to be tested, and the external loudspeaker is provided with a third controller used for receiving the test starting instruction in the voice performance test method to control the external loudspeaker to play test audio.

By adopting the technical scheme, the voice performance testing device comprises an equipment bearing wheel provided with a plurality of equipment fixing components, a plurality of devices to be tested are fixed on the equipment bearing wheel conveniently, so that the voice performance testing device can test the voice performance of the devices to be tested at one time, the voice performance testing device further comprises a base with a motor, an output shaft of the motor is fixedly connected to the equipment bearing wheel coaxially, the motor can drive the equipment bearing wheel to rotate, the equipment bearing wheel is driven to rotate when one device to be tested completes the voice performance test so as to move the next device to be tested to a preset target position for testing, an external loudspeaker is arranged above the equipment bearing wheel and used for playing audio required by the voice test, a signal transmitter used for transmitting signals with the devices to be tested is further arranged on the equipment bearing wheel, so that a test instruction is transmitted to the devices to be tested to control the devices to be tested to play/record The operation such as audio frequency reduces the workload of workers; therefore, the voice performance testing device can automatically and continuously test a plurality of devices to be tested, and the testing efficiency is improved.

The third objective of the present application is achieved by adopting the following technical scheme:

a voice performance testing system comprising:

the audio playing and recording module is used for acquiring the test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

the audio processing module is used for acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model and generating an algorithm processing signal;

the voice performance evaluation module is used for acquiring the microphone acquisition signal and the algorithm processing signal, and inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

and the test result generation module is used for comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

By adopting the technical scheme, the test item information is obtained so as to obtain the loudspeaker and the audio frequency which are required to be used by the next test item from the test item information, when the test is started, an audio playing instruction is generated based on the loudspeaker and the audio frequency which are required to be used by the test item and is sent to the external loudspeaker and the built-in loudspeaker of the equipment to be tested, an audio recording instruction is generated and is sent to the built-in microphone of the equipment to be tested, so that the equipment to be tested plays the test audio frequency, and a microphone collecting signal is recorded; inputting signals collected by a microphone into an audio processing model so as to obtain algorithm processing signals processed by various algorithms in the audio processing model; the microphone acquisition signal and the algorithm processing signal are input into a built-in voice performance evaluation model of the equipment to be tested for comparison and analysis so as to obtain a voice performance index reflecting the voice processing performance of the audio processing model, the microphone acquisition signal and the algorithm processing signal acquired by the equipment to be tested are not required to be transmitted to an electronic computer for testing through professional voice performance testing software, the purchase cost of the professional voice performance testing equipment and the professional voice performance testing software is saved, and the problem of low testing efficiency caused by compatibility is solved; and comparing the voice performance index with a preset voice performance threshold value to generate a voice performance test result for evaluating the voice processing performance of the audio processing model.

The fourth objective of the present application is achieved by adopting the following technical scheme:

a computer device comprising a memory, a processor and a computer program stored in said memory and executable on said processor, said processor implementing the steps of the above-mentioned speech performance testing method when executing said computer program.

The fifth aim of the invention is realized by adopting the following technical scheme:

a computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the above-mentioned speech performance testing method.

In summary, the present application includes at least one of the following beneficial technical effects:

1. the method has the advantages that the application program for performing the voice performance test is built in the device to be tested, and the microphone acquisition signal recorded by the device to be tested is compared with the algorithm processing signal processed by the audio processing model, so that a voice performance test result is generated, the effect that a computer provided with professional voice performance test software is not needed when the voice performance test is performed is achieved, the performance of the device to be tested is fully exerted, and the cost of purchasing the professional voice performance test device and the professional voice performance test software by an enterprise is reduced.

2. And creating an energy value waveform display window, and displaying the energy value waveforms of the microphone acquisition signal and the algorithm processing signal in the energy value waveform display window, so that a tester can compare the difference between the microphone acquisition signal recorded by the equipment to be tested and the algorithm processing signal processed by the audio processing model in real time, and problems occurring in the voice performance test process can be eliminated in time, and the detection accuracy is improved.

3. Through the speech performance testing device disclosed by the application, the staff can fix a plurality of devices to be tested on the equipment bearing wheel at one time, the speech performance testing device can automatically test the speech performance of the devices to be tested, and meanwhile, the next device to be tested is automatically replaced to test after the test of one device to be tested is completed, so that the working strength of the testing staff is reduced, and the detection efficiency is improved.

Drawings

FIG. 1 is a flow chart of a method for testing speech performance according to an embodiment of the present application;

FIG. 2 is a flowchart illustrating an implementation of step S30 in the voice performance testing method according to an embodiment of the present application;

FIG. 3 is a flowchart illustrating another implementation of step S30 in the voice performance testing method according to an embodiment of the present application;

FIG. 4 is a flowchart illustrating another implementation of step S30 in the voice performance testing method according to an embodiment of the present application;

FIG. 5 is a flowchart illustrating another implementation of step S30 in the voice performance testing method according to an embodiment of the present application;

FIG. 6 is a flowchart illustrating another implementation of step S30 in the voice performance testing method according to an embodiment of the present application;

FIG. 7 is a flowchart illustrating an implementation of step S10 in the method for testing speech performance according to an embodiment of the present application;

FIG. 8 is a flow chart of another implementation of the method for testing speech performance according to an embodiment of the present application;

FIG. 9 is a schematic structural diagram of a speech performance testing apparatus according to an embodiment of the present application;

FIG. 10 is a cross-sectional view taken along line A-A of FIG. 9;

fig. 11 is a partially enlarged view of a portion a in fig. 10;

FIG. 12 is a functional block diagram of a speech performance testing system according to an embodiment of the present application;

fig. 13 is a schematic diagram of an apparatus in an embodiment of the present application.

Description of reference numerals:

1. a base; 11. a motor; 12. an output shaft; 2. an equipment carrying wheel; 21. a device securing assembly; 22. a signal transmitter; 23. a connecting portion; 24. an equipment accommodating groove; 211. an equipment clamping assembly; 212. an elastic member; 213. a clamping block; 214. positioning holes; 2131. positioning a rod; 2132. a grip portion; 3. an external speaker; 4. a device to be tested; 5. a sound-insulating housing; 51. a sliding door; 52. a test bin; 53. and (5) loading and unloading the bin by the equipment.

Detailed Description

The present application is described in further detail below with reference to the attached drawings.

In an embodiment, as shown in fig. 1, the present application discloses a voice performance testing method, which specifically includes the following steps:

s10: and acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested.

In this embodiment, the test item information refers to a voice performance test item required by the device to be tested, where the test item includes a noise reduction test, an echo cancellation test, and a duplex echo cancellation test; the external loudspeaker is a loudspeaker on an external player for playing test audio, and the built-in loudspeaker is a loudspeaker carried by the device to be tested.

Specifically, the noise reduction test requires an external speaker to play a noise source, and the target voice can be played through the external speaker or an internal speaker; the echo cancellation test needs a built-in loudspeaker to play a test sound source; the duplex echo cancellation test requires a built-in speaker and an external speaker to play the test audio.

Specifically, because the test audio required by different test items may be different, and the number and types of speakers required to play audio may also be different, it is necessary to obtain the speaker and audio required to be used by the next test item from the test item information, generate an audio playing instruction based on the speaker and audio required to be used by the next test item, and send the audio playing instruction to the external speaker and the built-in speaker of the device under test, so as to perform a corresponding voice test item; and generating an audio recording instruction while sending the audio playing instruction, so that the equipment to be tested records the microphone acquisition signal to perform subsequent audio processing and comparison.

Specifically, the test audio can use a uniformly quantized multi-band audio source, wherein the frequency of the waveform is selected from 32Hz sine wave, 64Hz sawtooth wave, 125Hz square wave, 125Hz sine wave, 500Hz sawtooth wave, 1KHz square wave, 2KHz sine wave, 4KHz sawtooth wave, 8KHz square wave and 16KHz sine wave.

S20: and acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal.

In this embodiment, the microphone acquisition signal refers to an acoustic signal recorded by a microphone of the device to be tested during a voice performance test, and the microphone acquisition signal is derived from an audio played by an external speaker and/or an audio played by a built-in speaker of the device to be tested; the audio processing model refers to a processing model for processing the microphone collecting signal so as to process the voice audio collected by the microphone to obtain voice audio which is easier to be recognized by people.

Specifically, the audio processing model includes a noise reduction algorithm, an echo cancellation algorithm, and a duplex echo cancellation algorithm, and is configured to process a microphone acquisition signal recorded by the device under test to generate an algorithm processing signal with noise and echo filtered, and to perform voice performance evaluation on the algorithm processing signal subsequently to obtain an audio processing effect of the audio processing model, where the algorithm processing signal includes a noise reduction algorithm processing signal, an echo cancellation algorithm processing signal, and a duplex algorithm processing signal.

S30: and acquiring a microphone acquisition signal and an algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation, and obtaining a voice performance index.

In this embodiment, the speech performance evaluation model includes a noise reduction evaluation algorithm, an echo cancellation evaluation algorithm and a duplex echo cancellation evaluation algorithm; the voice performance index refers to voice performance parameters obtained by inputting a microphone acquisition signal into a voice performance evaluation model and calculating the voice performance parameters, and comprises a voice suppression ratio, a noise suppression ratio, an echo return loss gain value, a duplex voice suppression ratio and a duplex echo suppression ratio.

Specifically, after the voice performance evaluation model is created, the voice performance evaluation model is made into an application program, the voice performance evaluation model is installed in the equipment to be tested before the voice performance test is carried out, so that the microphone acquisition signal and the algorithm processing signal are evaluated on the equipment to be tested, compared with the traditional voice performance test in which the audio related to the test needs to be sent to an electronic computer provided with professional voice performance test software for evaluation, the purchase cost of the professional voice performance test equipment and the professional voice performance test software is reduced, and the compatibility problem caused by the transfer of the audio related to the test among different equipment is solved.

S40: and comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

In this embodiment, the speech performance threshold refers to a threshold of each speech performance parameter used for comparing with the speech performance index to determine whether the algorithm processing signal can satisfy the qualified standard.

In particular, the voice performance threshold includes a noise suppression threshold, a voice suppression threshold, an echo return loss gain threshold, a duplex echo suppression threshold, and a duplex voice suppression threshold.

Specifically, for the noise reduction test, when the average noise suppression ratio is greater than the noise suppression threshold and the average voice suppression ratio is less than the voice suppression threshold, the voice noise reduction is considered to meet the index requirement; otherwise, the index requirement is not met; for the echo cancellation test, if the echo return loss gain value is smaller than the echo return loss gain threshold value, the echo cancellation is considered to be abnormal, and if the echo return loss gain value is larger than the echo return loss gain threshold value, the echo cancellation is considered to be successful; for the duplex echo cancellation test, when the average duplex echo suppression ratio is greater than the duplex echo suppression threshold value and the average duplex voice suppression ratio is less than the duplex voice suppression threshold value, the duplex echo cancellation test is considered to meet the index requirement; otherwise, the index requirement is not met; and if the noise reduction test, the echo cancellation test and the duplex echo cancellation test of the equipment to be tested all meet the index requirements, generating a qualified voice performance test result, otherwise, generating a unqualified voice performance test result.

In the embodiment, test item information is acquired so as to obtain a speaker and audio needed by the next test item from the test item information, when a test is started, an audio playing instruction is generated based on the speaker and audio needed by the test item and is sent to an external speaker and a built-in speaker of the device to be tested, an audio recording instruction is generated and is sent to a built-in microphone of the device to be tested, so that the device to be tested plays a test audio, and a microphone collecting signal is recorded; inputting signals collected by a microphone into an audio processing model so as to obtain algorithm processing signals processed by various algorithms in the audio processing model; the microphone acquisition signal and the algorithm processing signal are input into a built-in voice performance evaluation model of the equipment to be tested for comparison and analysis so as to obtain a voice performance index reflecting the voice processing performance of the audio processing model, the microphone acquisition signal and the algorithm processing signal acquired by the equipment to be tested are not required to be transmitted to an electronic computer for testing through professional voice performance testing software, the purchase cost of the professional voice performance testing equipment and the professional voice performance testing software is saved, and the problem of low testing efficiency caused by compatibility is solved; and comparing the voice performance index with a preset voice performance threshold value to generate a voice performance test result for evaluating the voice processing performance of the audio processing model.

In one embodiment, as shown in fig. 2, before step S30, the method further includes the steps of:

s31: and acquiring a noise reduction evaluation algorithm, an echo cancellation evaluation algorithm and a duplex echo cancellation evaluation algorithm, and creating a voice performance evaluation model based on the noise reduction evaluation algorithm, the echo cancellation evaluation algorithm and the duplex echo cancellation evaluation algorithm.

Specifically, the noise reduction evaluation algorithm aligns phases of a microphone acquisition signal and a noise reduction algorithm processing signal, and calculates a voice suppression ratio and a noise suppression ratio as voice performance indexes; the echo cancellation evaluation algorithm aligns phases of a microphone acquisition signal and an echo cancellation algorithm processing signal, and calculates an echo return loss gain value as a voice performance index; the duplex echo cancellation evaluation algorithm aligns the acquired microphone acquisition signal and the duplex algorithm processing signal respectively, takes the difference value of the duplex algorithm processing signal and the microphone acquisition signal as an algorithm evaluation signal, aligns the microphone acquisition signal and the algorithm evaluation signal, and calculates a duplex voice suppression ratio and a duplex echo suppression ratio as voice performance indexes.

S32: and sending the speech performance evaluation model to the equipment to be tested.

Specifically, after the voice performance evaluation model is created, the application program with the built-in voice performance evaluation model is installed in the device to be tested before the voice performance test is started, so that the device to be tested is used for running the application program to obtain the voice performance index, and then the voice performance test result is generated, the purchase cost of the voice performance test device is saved, and meanwhile, the performance of hardware of the device to be tested is fully utilized.

Further, after the voice performance test is completed and before the equipment leaves the factory, the application program with the built-in voice performance evaluation model can be uninstalled, so that the memory occupation of the equipment leaving the factory is reduced, and the use experience of a user is improved.

In one embodiment, as shown in fig. 3, in step S30, the method further includes the steps of:

s33: and creating an energy value waveform display window, displaying the energy value waveforms of the microphone acquisition signal and the algorithm processing signal in the energy value waveform display window, and comparing the waveforms of the microphone acquisition signal and the algorithm processing signal in real time.

In this embodiment, the energy value waveform display window refers to a program capable of displaying the energy value waveform of the acoustic signal in the form of an image on the user interface.

Specifically, the energy value waveform display window can display the energy value waveforms of the microphone collecting signals and the algorithm processing signals on a screen of the device to be tested, so that a tester can observe the test process in real time, and problems occurring in the voice performance test process can be eliminated in time.

Further, the display content of the energy value waveform display window can be transmitted to other equipment, so that a tester can observe the energy value waveform diagram of the test process in a different place.

In an embodiment, as shown in fig. 4, in step S30, the method further includes the steps of:

s34: acquiring a microphone acquisition signal and a noise reduction algorithm processing signal, aligning the microphone acquisition signal and the noise reduction algorithm processing signal, and calculating a voice suppression ratio and a noise suppression ratio as voice performance indexes.

Specifically, the specific flow of the noise reduction evaluation algorithm is as follows: the built-in loudspeaker plays a section of target voice signal, and the external loudspeaker plays a section of noise signal; the method comprises the steps of recording a microphone acquisition signal by a built-in microphone, carrying out noise reduction processing on the microphone acquisition signal by using a noise reduction algorithm to obtain a noise reduction algorithm processing signal, aligning the microphone acquisition signal with the noise reduction algorithm processing signal, and calculating a voice suppression ratio and a noise suppression ratio.

Specifically, the calculation formula of the Speech Suppression ratio (Speech Suppression Rate, SSR) is:

the Noise Suppression ratio (Noise Suppression Rate, NSR) is calculated by the following formula:

in one embodiment, as shown in fig. 5, in step S30, the method further includes the steps of:

s35: acquiring a microphone acquisition signal and an echo cancellation algorithm processing signal, aligning the microphone acquisition signal and the echo cancellation algorithm processing signal, and calculating an echo return loss gain value as a voice performance index.

Specifically, the specific flow of the echo cancellation evaluation algorithm is as follows: playing a section of voice signal by the built-in loudspeaker, acquiring a microphone acquisition signal by the built-in microphone, carrying out echo cancellation processing on the microphone acquisition signal by using an echo cancellation algorithm to obtain an echo cancellation algorithm processing signal, aligning the microphone acquisition signal d (k) with the echo cancellation algorithm processing signal e (k), and calculating an echo return loss gain value.

Specifically, the Echo Return Loss gain (ERLE) is calculated by the following formula:

in an embodiment, as shown in fig. 6, in step S30, the method further includes the steps of:

s36: respectively acquiring a microphone acquisition signal and a duplex algorithm processing signal, aligning the microphone acquisition signal and the duplex algorithm processing signal, subtracting the microphone acquisition signal from the duplex algorithm processing signal to obtain an algorithm evaluation signal, aligning the microphone acquisition signal and the algorithm evaluation signal, and calculating a duplex voice suppression ratio and a duplex echo suppression ratio as voice performance indexes.

Specifically, the specific flow of the duplex echo cancellation evaluation algorithm is as follows: firstly, when an external loudspeaker plays a section of target voice, a microphone acquisition signal is x, then, the external loudspeaker plays the same section of target voice, and when an internal loudspeaker plays another section of voice signal, the internal microphone records an external voice signal and processes the external voice signal by using a duplex echo cancellation algorithm to obtain a duplex algorithm processing signal y, aligns the x and y signals, subtracts the x signal from the y to obtain an algorithm evaluation signal z, aligns the microphone acquisition signal x and the algorithm evaluation signal z, and calculates a duplex evaluation index, wherein the duplex evaluation index comprises a duplex voice suppression ratio and a duplex echo suppression ratio.

Specifically, the calculation formula of the duplex voice suppression ratio is as follows:

the calculation formula of the duplex echo suppression ratio is as follows:

in one embodiment, as shown in fig. 7, before step S10, the method further includes the steps of:

s11: and acquiring the test serial number and the position information of the equipment to be tested, and sending a test preparation instruction to the motor based on the test serial number and the position information of the target equipment to be tested.

In this embodiment, the test serial number refers to a serial number assigned to a plurality of devices to be tested placed on the device bearing wheel, and is used for determining a test sequence of the devices to be tested; the position information is information in which the current position of each device under test is recorded.

Specifically, a test serial number and position information of the device to be tested are acquired, a next target device to be tested is determined and marked based on the test serial number of the device to be tested, and relative position information of the current position of the target device to be tested and a preset target position is calculated based on the position information of the target device to be tested, so that action parameters of a motor are generated based on the relative position information, and the target device to be tested is moved to the preset target position to perform voice performance test.

S12: and acquiring the position information of the target equipment to be tested in real time, and sending a test starting instruction to the equipment bearing wheel and the external loudspeaker when the position information of the target equipment to be tested is successfully matched with the preset target position.

In this embodiment, the preset target position refers to an optimal position for the device under test to perform the voice performance test.

Specifically, the position information of the target equipment to be tested is obtained in real time, so that the relative position information of the position of the target equipment to be tested and a preset target position is obtained, the rotating angle of the motor is controlled in a closed loop mode, the accuracy of the rotating position of the equipment bearing wheel is improved, and the test is started when the position information of the target equipment to be tested is successfully matched with the preset target position.

In an embodiment, as shown in fig. 8, after step S40, the method further includes the steps of:

s50: and after the target equipment to be tested is tested, marking the next target equipment to be tested from the equipment to be tested based on the test serial number of the tested target equipment to be tested.

Specifically, after a target device under test completes a voice performance test, a next target device under test is selected and marked based on a test serial number of the target device under test that completes the test.

It should be understood that, the sequence numbers of the steps in the foregoing embodiments do not imply an execution sequence, and the execution sequence of each process should be determined by its function and inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

In an embodiment, as shown in fig. 9, the voice performance testing apparatus includes a base 1, an equipment carrying wheel 2, and an external speaker 3, where the base 1 is provided with a motor 11 for driving the equipment carrying wheel 2 to rotate, an output shaft 12 of the motor 11 is fixedly connected to the equipment carrying wheel 2 so that the output shaft 12 of the motor 11 and the equipment carrying wheel 2 rotate coaxially, and the motor 11 is provided with a first controller for receiving a preparation test command in the voice performance testing method to control the output shaft 12 of the motor 11 to rotate, so that the motor 11 rotates a target device 4 to be tested to a preset target position based on an action parameter in the preparation test command.

As shown in fig. 10 and 11, the device bearing wheel 2 includes a connecting portion 23 for connecting the output shaft 12 and a plurality of device fixing assemblies 21 for fixing the device to be tested 4 to the device bearing wheel 2, one end of the device fixing assembly 21 is fixedly connected to the connecting portion 23, one end of the device fixing assembly 21 away from the connecting portion 23 is provided with a device accommodating groove 24 for accommodating the device to be tested 4, two mutually symmetrical side walls of the device accommodating groove 24 are provided with two device clamping assemblies 211, each device clamping assembly 211 includes an elastic member 212 and a clamping block 213, one end of the elastic member 212 is fixedly connected to a side wall of the device accommodating groove 24, the other end of the elastic member 212 is fixedly connected to the clamping block 213, and the device fixing assembly 21 plays a role of clamping the device to be tested 4 through the two device clamping assemblies 211; one side that grip block 213 is connected with elastic component 212 is provided with locating lever 2131, and the equipment holding groove 24 is used for two lateral walls of fixed elastic component 212 all to have seted up locating hole 214, and locating lever 2131 wears to establish and sliding connection in locating hole 214 to the direction of motion of restriction grip block 213 makes grip block 213 distribute more evenly to the clamping-force of awaiting measuring equipment 4.

Preferably, a holding portion 2132 is disposed on a side, away from the base 1, of the clamping block 213, so that when the thickness of the device to be tested 4 is large, a worker can move the clamping block 213 to take the device to be tested 4 conveniently, and the clamping block 213 can be made of a soft material, so as to reduce the probability that the clamping block 213 wears the device to be tested 4 when abutting against the device to be tested 4.

As shown in fig. 11, the connection portion 23 is provided with a signal transmitter 22 for transmitting signals between the device carrying wheel 2 and the device under test 4, and the signal transmitter 22 is connected to the device under test 4, on one hand, for transmitting a test starting instruction to the device under test 4, and on the other hand, for acquiring a test result after the voice performance test is completed; the equipment bearing wheel 2 is provided with a second controller for receiving a test starting instruction in the voice performance test method so as to control the signal transmitter 22 to send the test starting instruction to the equipment to be tested 4; in this embodiment, the signal transmitter 22 is a data line, and in actual use, communication may be established between the device carrier wheel 2 and the device under test 4 through a wireless communication module such as a bluetooth module or a network connection module.

The device to be tested 4 is located in the device accommodating groove 24, one end of the device to be tested 4 is provided with a data interface, one end of the device to be tested 4, which is provided with the data interface, faces the connecting portion 23, one end of the device to be tested 4, which is far away from the connecting portion 23, abuts against the side wall of the device accommodating groove 24, which is far away from the connecting portion 23, and two side faces, which are adjacent to one end of the device to be tested 4, which are provided with the data interface, abut against the two clamping blocks 213 respectively.

As shown in fig. 9 and 10, the voice performance testing apparatus further includes a sound insulation housing 5 for separating the outside from the testing space of the device 4 to be tested, so that on one hand, the influence of external noise on the voice performance test of the device 4 to be tested can be reduced, and the accuracy of the voice test is improved, on the other hand, the influence of noise sources on the hearing of the staff during the noise reduction performance test of the device 4 to be tested can be reduced, and the comfort of the staff during the testing work can be improved; the sound insulation shell 5 is in a shape of a cylinder with two overlapped parts, the sound insulation shell 5 comprises two chambers, namely a test chamber 52 for testing the voice performance of the device to be tested 4 and a device loading and unloading chamber 53 for mounting the device to be tested 4 to the device bearing wheel 2 or taking the device to be tested 4 from the device bearing wheel 2, wherein the test chamber 52 is in a shape of a complete cylinder, an opening is formed between the test chamber 52 and the device loading and unloading chamber 53, and therefore the device to be tested 4 is transferred between the test chamber 52 and the device loading and unloading chamber 53 through the device bearing wheel 2.

An opening for placing the device under test 4 into or taking the voice performance testing device out is formed in one side of the device loading and unloading bin 53, which is far away from the testing bin 52, a sliding door 51 for opening or closing the opening is further arranged at the opening, the external loudspeaker 3 is fixedly connected to the top of the testing bin 52, and the external loudspeaker 3 is provided with a third controller for receiving a test starting instruction in the voice performance testing method to control the external loudspeaker 3 to play testing audio.

In an embodiment, a voice performance testing system is provided, which corresponds to the voice performance testing methods in the above embodiments one to one.

As shown in fig. 12, a voice performance testing system includes an audio playing and recording module, an audio processing module, a voice performance evaluating module, and a testing result generating module. The detailed description of each functional module is as follows: the audio playing and recording module is used for acquiring the test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

the audio processing module is used for acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into the audio processing model and generating an algorithm processing signal;

the voice performance evaluation module is used for acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into the voice performance evaluation model for local evaluation to obtain a voice performance index; and the test result generation module is used for comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

For the specific definition of the voice performance testing system, reference may be made to the above definition of the voice performance testing method, which is not described herein again; all modules in the voice performance test system can be completely or partially realized through software, hardware and a combination of the software and the hardware; the modules can be embedded in a hardware form or independent from a processor in the computer device, or can be stored in a memory in the computer device in a software form, so that the processor can call and execute operations corresponding to the modules.

In one embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as shown in fig. 13. The computer device includes a processor, a memory, a network interface, and a database connected by a system bus. Wherein the processor of the computer device is configured to provide computing and control capabilities. The memory of the computer device comprises a nonvolatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of an operating system and computer programs in the non-volatile storage medium. The database of the computer equipment is used for storing data such as test audio, target voice, an audio processing model, a voice performance evaluation model, voice performance indexes, microphone acquisition signals, algorithm processing signals and the like. The network interface of the computer device is used for communicating with an external terminal through a network connection. The computer program is executed by a processor to implement a method of speech performance testing.

In one embodiment, there is provided a computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the following steps when executing the computer program:

s10: acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

s20: acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal;

s30: acquiring a microphone acquisition signal and an algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

s40: and comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

In one embodiment, a computer-readable storage medium is provided, having a computer program stored thereon, which when executed by a processor, performs the steps of:

s10: acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

s20: acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal;

s30: acquiring a microphone acquisition signal and an algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

s40: and comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the embodiments provided herein may include non-volatile and/or volatile memory, among others. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions.

The above-mentioned embodiments are only used for illustrating the technical solutions of the present application, and not for limiting the same; although the present application has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art; the technical solutions described in the foregoing embodiments may still be modified, or some features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present application and are intended to be included within the scope of the present application.

Claims (10)

1. A voice performance testing method is characterized in that: the voice performance testing method comprises the following steps:

acquiring test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model, and generating an algorithm processing signal;

acquiring the microphone acquisition signal and the algorithm processing signal, and inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

and comparing the voice performance index with a voice performance threshold value to generate a voice performance test result.

2. The method for testing speech performance of claim 1, wherein: before the step of obtaining the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation, and obtaining a voice performance index, the method further comprises the following steps:

acquiring a noise reduction evaluation algorithm, an echo cancellation evaluation algorithm and a duplex echo cancellation evaluation algorithm, and creating a voice performance evaluation model based on the noise reduction evaluation algorithm, the echo cancellation evaluation algorithm and the duplex echo cancellation evaluation algorithm;

and sending the voice performance evaluation model to the equipment to be tested.

3. The method for testing speech performance of claim 1, wherein: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and creating an energy value waveform display window, displaying energy value waveforms of the microphone acquisition signal and the algorithm processing signal in the energy value waveform display window, and comparing the waveforms of the microphone acquisition signal and the algorithm processing signal in real time.

4. The method of claim 2, wherein: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and acquiring the microphone acquisition signal and the noise reduction algorithm processing signal, aligning the microphone acquisition signal and the noise reduction algorithm processing signal, and calculating a voice suppression ratio and a noise suppression ratio as voice performance indexes.

5. The method of claim 2, wherein: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

and acquiring the microphone acquisition signal and the echo cancellation algorithm processing signal, aligning the microphone acquisition signal and the echo cancellation algorithm processing signal, and calculating an echo return loss gain value as a voice performance index.

6. The method of claim 2, wherein: the method comprises the steps of acquiring the microphone acquisition signal and the algorithm processing signal, inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index, and further comprises the following steps:

respectively acquiring the microphone acquisition signal and the duplex algorithm processing signal, aligning the microphone acquisition signal and the duplex algorithm processing signal, subtracting the microphone acquisition signal from the duplex algorithm processing signal to obtain an algorithm evaluation signal, aligning the microphone acquisition signal and the algorithm evaluation signal, and calculating a duplex voice suppression ratio and a duplex echo suppression ratio as voice performance indexes.

7. The method for testing speech performance of claim 1, wherein: the method comprises the following steps of obtaining test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the device to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the device to be tested, and the method also comprises the following steps:

acquiring a test serial number and position information of equipment to be tested, and sending a test preparation instruction to the motor based on the test serial number and the position information of target equipment to be tested;

the method comprises the steps of obtaining position information of target equipment to be tested in real time, and sending a test starting instruction to an equipment bearing wheel and an external loudspeaker when the position information of the target equipment to be tested is successfully matched with a preset target position.

8. The method for testing speech performance of claim 1, wherein: after the step of comparing the voice performance index with the voice performance threshold value and generating a voice performance test result, the method also comprises the following steps:

and after the target equipment to be tested is tested, marking the next target equipment to be tested from the equipment to be tested based on the test serial number of the tested target equipment to be tested.

9. A voice performance testing device is characterized in that: the device comprises a base (1), a device bearing wheel (2) and an external loudspeaker (3), wherein a motor (11) is arranged on the base (1), an output shaft (12) of the motor (11) is coaxially and fixedly connected to the device bearing wheel (2), a plurality of device fixing components (21) used for fixing a device to be tested (4) on the device bearing wheel (2) are arranged on the device bearing wheel (2), a signal transmitter (22) used for transmitting signals with the device to be tested (4) mutually is further arranged on the device bearing wheel (2), and the external loudspeaker (3) is arranged above the device bearing wheel (2); the motor (11) is provided with a first controller for receiving a test preparation command in the voice performance test method of claim 7 to control the rotation of the output shaft (12) of the motor (11), the equipment bearing wheel (2) is provided with a second controller for receiving a test start command in the voice performance test method of claim 7 to control the signal transmitter (22) to send the test start command to the target equipment under test (4), and the external loudspeaker (3) is provided with a third controller for receiving the test start command in the voice performance test method of claim 7 to control the external loudspeaker (3) to play test audio.

10. A speech performance testing system, comprising:

the audio playing and recording module is used for acquiring the test item information, generating an audio playing instruction based on the test item information, sending the audio playing instruction to an external loudspeaker and a built-in loudspeaker of the equipment to be tested, and simultaneously generating an audio recording instruction and sending the audio recording instruction to a built-in microphone of the equipment to be tested;

the audio processing module is used for acquiring a microphone acquisition signal recorded by a built-in microphone, inputting the microphone acquisition signal into an audio processing model and generating an algorithm processing signal;

the voice performance evaluation module is used for acquiring the microphone acquisition signal and the algorithm processing signal, and inputting the microphone acquisition signal and the algorithm processing signal into a voice performance evaluation model for local evaluation to obtain a voice performance index;

and the test result generation module is used for comparing the voice performance index with the voice performance threshold value to generate a voice performance test result.

Images